Spouse bereavement is a complex process that involves intense personal pain. Although it has been recognized as a major public health problem, little is known about the impact illness has on spouses prior to patients' death. The overall purpose of this study is to compare data obtained from spouses before the patients' death with data obtained after the death. The aim is to document the impact of the illness and death on the survivor's health, their relationship with others, and the subsequent changes in their roles and activities. Prospective data will be collected at approximately six week intervals two months after the cancer diagnosis, and continue for a period of six months. Concurrent data will be obtained six weeks after the death of the patient and will continue for 25 months after death. Using a combination of standardized instruments and semi-structured interviews, data will be obtained from spouses on symptoms, health perception, personal and family relationships, bereavement, utilization of services, evaluation of health professionals support, and identify transition characteristics. Approximately 100 spouses will be followed. To date, prospective data is being collected on 95 spouses, and concurrent data collection has been initiated on 51 surviving spouses. Analysis will consist of comparing mean scores on prospective and concurrent variables in repeated measures univariate and multivariate analyses of variance. Correlations will be examined to develop a primary predictive model of spouses at risk for developing problems. Data obtained in this study will provide insight into our understanding of changes which surviving spouses experience that are associated with a life-threatening illness, and subsequent death of the patient. Those changes that are associated with increased health problems, loss of social relationships, and meaningful role changes will be identified.